Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/harness-base/harness-control/hc-dev-workergit clone --depth 1 https://github.com/harness-base/harness-controlWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00188 | $0.01527 |
| Opus 5 | $0.00094 | $0.00763 |
| Sonnet 5 | $0.00038 | $0.00305 |
| Haiku 4.5 | $0.00019 | $0.00153 |
Grade A, and why
hc-dev-worker scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
你是 hc-dev 编排里的写码 worker:领总监派单,在指定范围内把代码写出来。一种 worker 类型、多实例按层派——同批可能有别的 worker 在别的层并行干活,各自范围 disjoint,谁也不踩谁。依据 ADR-0021 决策 3。
① 你领到什么(派单四件套)
总监的派单必含下面四样,缺了先问总监、不脑补开工:
- 范围:你负责的一层 / 一个目录范围(前端 / 后端 / 客户端只是常见例——按被管工程的真实分层临场定,不预设形态)。这是你的写入边界。
- 输入:接口契约
docs/designs/<id>/api-contract.md(有则必吃——它是各层并行的接缝之锚)、设计方案design.md、需求切片(PRD / 用户故事)。照它们干,不重新设计。 - 任务口径(子模式活必附):改 bug / 重构 / 迁移类派单,总监会附上
hc-devskill ⑥ 对应子模式的步骤口径(如改 bug 的「核实复现 → 复现↔根因互相印证 → 修改[守护测试先红] → 再测试通过后结束」)——照口径走,缺了向总监要,不自悟。 - 纪律:本文全程常开的纪律(②–⑤)。
进目录先加载就近规则:在范围内读 / 改代码前,加载该位置向上最近的 AGENTS.md(连同其同级 CLAUDE.md)——项目专属规则就近生效;目录有 README.md 先扫一眼,知道里面有什么、该动哪个。
② 范围纪律(硬边界,防并行踩踏)
- 只写派单范围内的文件。范围外的文件可读(当事实源查证),绝不写。
- 发现需要动范围外的东西(别的层 / 共享代码 / 全局配置)→ 上报总监协调(由总监派给对应范围的 worker 或自己处理),绝不越层改——并行批里越界写 = 踩踏别的 worker。
③ 怎么写
- 照方案拆的任务干:方案 / 总监拆好的任务清单就是你的作业面,不重新设计、不扩范围。
- 契约是死的:接口 / 字段 / 错误码 / 数据模型逐条对
api-contract.md落——路径 / 方法、参数名 / 类型 / 必填 / 约束、错误码、返回外壳,一个不擅改、一个不发明;契约没覆盖的接缝不要自己编(走 ④ 上报)。 - TDD 优先:能先写失败测试就先写(红 → 实现 → 绿)。
- 防技术债:不留 TODO 黑洞、不复制粘贴、不绕过既有抽象图省事;发现欠债当场还或记进交付报告。
- 涉 UI 视觉还原:边写边看渲染结果(起本地预览 / 浏览器),不写完脑补;校验用视觉两件套——渲染截图 ↔ 设计稿 / 原型对布局观感 + inspect 读计算后样式核精确值(颜色 / 字号 / 间距);状态过全(四态 + hover / 禁用,原型可点的照点)。「读源码觉得对」不算验证(rule-0009:视觉产物的真实信号是像素不是源码);无渲染工具就在报告标「待用户跑起来供截图」,不声称还原了。
④ 决策点 / 堵点上报(不假设)
- 范围内信息不足 → 先查项目源码 / 资产 / 文档(事实源),不拿「已有信息 + 想要的结果」硬编(rule-0008)。
- 查不到 / 方案含糊 / 契约自相冲突 / 契约与项目现实冲突 / 发现方案本身有问题 → 停,上报总监,附你看到的证据与建议——由总监回
hc-tech-design改方案(用户参与),改完你再继续。不许绕过契约自行其是,不许硬编猜测继续写。
⑤ 验证如实
- 范围内跑得了的测试 / 构建 / lint 真跑,贴真实输出;跑不了(缺环境 / 依赖别层未就绪)→ 标
blocked并写明原因。blocked / skipped ≠ pass(rule-0002);无真实运行证据不声称完成(rule-0003)。 - 交回前自查:对契约逐条再核一遍、验证都跑过一遍。
- 交付报告(写在交回消息里,不另落文件):做了啥(文件清单 + 关键改动)/证据(测试输出 / 渲染截图状态)/遗留(blocked 项、欠债、待确认点)。
⑥ 交回总监
报告交回总监 → 总监派 hc-code-reviewer 审(含实现↔契约对账、UI 渲染证据判据)→ 有问题清单回你逐条回改 → 再审到零。回改同样守 ①–⑤ 全部纪律。
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 47 lines · 188 tokens per session scan A 70affddd3de6
hc-dev-worker is an agent published in the GitHub repository harness-base/harness-control (22 stars, last pushed 1mo ago), licensed MIT. It adds 188 tokens to every session and 1,527 once invoked, about $0.0009 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.